Project 1 Understanding the Complex Mechanisms Underlying Tumor Host Interactions and the Modulation of Tumor Angiogenesis by Cytokines and Gene Regulation. Aim 1 - The identification of early genetic mechanisms underlying the inhibitory effects of antiangiogenic agents on human endothelial cells and characterizing the role played by DOC1. Our previous work using cDNA microarray analysis, demonstrated that the expression of 4 genes DOC1, KLF4 and TC-1 were rapidly modulated by a variety of angiogenesis inhibitors and that DOC1 appeared to be an upstream regulator of KLF4 and TC-1. The function of DOC1, however, was completely unknown. Based on these data, we hypothesized that DOC1 may mediate some of the anti-angiogenic activity in endothelial cells by inducing apoptosis. We have developed a mouse monoclonal DOC1 antibody, which allowed us to detect endogenous DOC1 protein in human tissues for the first time. We have also demonstrated that DOC1 protein is expressed in the cytoplasm, on the membrane and in the nucleus of endothelial cells, and on the vasculature, stroma and muscularis in human cancer specimens. Aim 2 - Characterize the effects of the tumor-derived cytokine EMAP-II on endothelial cells and determine its mechanism of action. Endothelial Monocyte Activating Polypeptide-II (EMAP-II) is a proinflammatory cytokine that has potent effects on endothelial cells (ECs). It suppresses primary and metastatic tumor growth through its effects on tumor vasculature. Little is known about the molecular mechanisms by which EMAP-II exerts its antiangiogenic effects. We have demonstrated that when ECs are treated with EMAP-II, the EMAP-II enters the cytoplamsic compartment via &#945;5&#946;1 integrin receptors. We have identified PSMA7, a component of the 26S proteasome complex, as a cytoplasmic binding partner of EMAP-II. Binding of EMAP-II to PSMA7 via unique binding sites potentiates the interaction of PSMA7 with hypoxia-inducible factor-1 alpha (HIF-1&#945;) and increases the degradation of HIF-1&#945;. EMAP-II treatment of ECs resulted in decreased HIF-1&#945; protein and also inhibited HIF-1&#945; driven transcriptional activity. Downregulation of PSMA7 with SiRNA confirmed specific involvement of PSMA7 in EMAP-II mediated HIF-1&#945; degradation. This effect results in an inhibition of EC tube formation. Aim 3 - Modulation of tumor-host interactions and tumor growth using pathway targeted therapy and the development of models to better understand these effects. In order to better understand the processes involved in tumor angiogenesis we set out to develop transgenic mouse models with spontaneous tumors, mimicking tumor development in the patients we are treating. Successful development of such mouse models will allow us to better understand mechanisms involved in tumorigenesis, and to test therapies we are developing during different stages of tumor progression. Utilizing the cre-lox system, we were able to inactivate tumor suppressor genes in a tissue specific manner. The two tumor suppressors we focused on were the Multiple endocrine neoplasia type 1 (MEN1) and the von Hippel-Lindau (VHL) genes. Both MEN1 and VHL syndromes result in neuroendocrine tumors in patients that we treat and as such, the development of models patterning these entities would be important to our work. These highly vascular tumors provide an ideal model for the study of tumor host interactions and tumor vasculature. We have successfully developed a mouse model of pancreatic neuroendocrine tumors by pancreas selective deletion of the Men1 gene. We are still in the process of analyzing the pancreas specific deletion of the Vhl gene in our mice for a phenotype. Our work will focus on identifying mechanisms involved in tumor formation and in evaluating targeted therapies in these models. Project 2 Translation of anti-angiogenic and anti-vascular therapies to the clinic and the development of surrogate methods for measuring their activity. Aim 1 - Approaches to anti-angiogenic and anti-vascular gene therapy. A major goal of our laboratory is to develop tumor vascular targeted therapies to deliver anti-vascular agents selectively to the tumor bed. Considering the various limitations of current vector systems, we evaluated targeted AAVP, a hybrid prokaryotic/eukaryotic vector to deliver tumor necrosis factor-alpha (TNF-&#945;) to tumor vasculature. This bacteriophage vector has no native tropism for mammalian cells; however it targets gene products to tumor vasculature by using an alpha V beta 3 integrin ligand (termed RGD-4C) motif. We evaluated targeted and non-targeted AAVP vector expressing TNF-&#945;, EMAP-II and DOC1 in-vitro and in-vivo in human melanomas, prostate cancer and our transgenic pancreas neuroendocrine model. In mouse xenograft tumor models, AAVP-TNF-&#945; targeted tumor vasculature when delivered systemically, with no detectable virus in normal organs. Systemic delivery of targeted AAVP-TNF-&#945;, resulted in significant tumor reduction compared to empty AAVP and non-targeted AAVP-TNF-&#945;. Similar results were seen for EMAP-II and DOC1 delivery. In order to examine safety and efficacy in a clinically relevant large animal model, tumor vasculature targeting and efficacy of this vector was evaluated in companion dogs in collaboration with the NCI comparative oncology program. We evaluated AAVP-TNF-alpha in a total of 42 dogs with spontaneous tumors. A single systemic administration of AAVP-TNF-alpha in doses as high as 1x1013 virus particles did not show any significant toxicity. Four to eight repeated administrations of 5x1012 virus particles confirmed that AAVP is safe to administer systemically. We detected presence of AAVP in tumor vasculature as early as 4 - 6 hrs after administration as well as at day 4 after single administration of virus, with no virus seen in the normal tissues tested. Targeted delivery of AAVP- TNF showed expression of TNF-alpha by TaqMan RT-PCR. A total of 15 dogs received a minimum of 4 cycles of AAVP-TNF over a period of 4 weeks. We evaluated responses using RECIST criteria. Out of 15 evaluable dogs, two had significant responses by RECIST. Our future plans are to conduct a phase I clinical trial in patients using AAVP-TNF as well as to continue the development of AAVP-EMAP-II and AAVP-DOC1. Aim 2 - Imaging of tumor associated vasculature and the development of means to quantify changes in order to monitor the effects of vascular targeted therapy. A major focus of our laboratory investigation has been to develop methods with which to better analyze vascular structures and the continuum of angiogenesis in order to monitor the effects of the therapies we are developing. These analyses can be observer-dependent and lack standardization. Using the transgenic mouse model of pancreatic neuroendocrine tumor formation that we have developed in our laboratory, we have been studying improved methods for image guided vessel quantification and analysis. These neuroendocrine tumors are highly vascular and the vessels appear to become larger and more chaotic as the tumors grow. This model is ideal for the application of image quantification techniques in order to better characterize angiogenesis. These quantification techniques may be useful in the clinical setting and our future work will focus on applying these techniques to computed tomography or magnetic resonance angiography for noninvasive assessment of tumor vasculature. Information gained from this kind of analysis may be helpful for prognosis and evaluating treatment effects